On Tue, 11 Jul 2006, Doug Cutting wrote:
Probably this would get fixed more quickly if someone contributed a patch to
JavaCC. Even it were not committed, we could build our own version of
JavaCC. Any intrepid volunteers?
For patches that seem too kludgy to make it into Lucene's sources (for
example, to work around the lack of proper exception support under Windows
gcj in the query parser) a compromise could be to keep these patches in a
separate file and apply them to the Lucene sources before building them with
gcj. This is how PyLucene is built today.
Some patches have already been incorporated into the Lucene sources (for
example, in Searcher.java, to workaround gcj bug 15411).
Of course, the long term goal should be to no longer have any patches at all.
I've been working on PyLucene about two and a half years now and the number of
patches has remained fairly stable.
A nice side effect of trying to support gcj with Java Lucene by including it
into the Lucene test framework could be that the gcj developers might be more
inclined to taking a look at gcj-related issues that are thus made much easier
to reproduce.
Andi..
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
